Posts Tagged ‘kanji test’

v0.571 – and an extra experimental feature.

February 12, 2011 Comments off

Another update. This time I’m fixing some problems the previous update caused to users who create their own dictionaries. I would recommend updating even if you haven’t started a custom dictionary (majority of users), but if you have v0.57 already, you will only need to download the “executable only” archive of v0.571 (at the bottom of the zkanji download page).

I will now have to think about my next step. The decision is always difficult. I could start working on something new everyone benefits from, make some small feature that a few users have needed since ages now, or start creating a help file with pretty images, because a picture is worth 1000 words – or something like that.

Before I forget, if you like experimental stuff there is something for you! Close zkanji. I know that it’s always running. Open the zkanji.ini file in a plain text editor. Please don’t use SomeOffice or other monster programs because they can ruin stuff! Plain old notepad will do. Add the following two lines at the top of the ini file:


Save the ini file. (This why I prefer ini files over any kind of registry.)

Now start up zkanji and select a single kanji you like. Press Ctrl-B or click “BETA Test BETA” in the Tools menu. Voila, you can test your knowledge of the kanij strokes and stroke order!

As I said this is just experimental at the moment, so the kanji testing window is plain and ugly and there are controls you won’t know what they are for. You can uncheck “Show debug data” if you have the courage in you and select “Next stroke hint” if you need to know where to place the next stroke. There is nothing else to it. This test doesn’t record anything, doesn’t tell you what you did wrong and in general doesn’t do anything except for showing a red X or a green circle, depending on how your stroke turned out. If the stroke wasn’t correct, you have to draw it again.

The algorithm used in this experimental test is, as the name suggests, experimental too, so if you feel that it needs fine-tuning it’s because it needs fine-tuning. Unless I receive feedback, I won’t be able to fix it or do anything with it. It might not even get released in a nice and clean user friendly window in that case. So please, send your suggestions!

Some nice people I have shown this already told me, that the algorithm is too strict, and if they made some strokes that they felt should be passed as acceptable but was still a bit “non-standard”, the algorithm didn’t accept it. BUT. (There is always one, right?) I’ve never intended this test to accept so loose kanji like the working handwriting recognition window would. Why is that? Because this thing is made for seeing how accurately can a student reproduce a kanji. Please only complain, if you tried your best, and still can’t draw kanji. That best can be anything, I know that it is difficult to write with a mouse, so don’t be shy!